结合Gauss级数对时间序列中奇异值预测

Use the Gauss Series in Forecasting the Series Data with the Strange Points

时间序列数据奇异值的预测一直受到学界的关注。利用具有特殊性质的正交级数模型——Gauss模型,与时间序列模型相结合建立混合模型,能较好地预测具有奇异值的时间序列数据。并通过模型的实际运用,证明了模型的可行性和有效性。

To mix the Orthogonal Series Models-Gauss Series Model,and Time Series Model in the forecasting-model,so that it can estimate the clara that contains the strange points more accurately.The result is: this mixed model can forecast well and truly in short time.